REAL-TIME IOT MONITORING AND BRIX VALUE PREDICTION IN FOOD PROCESSING USING WEIGHT RATIO AND LINEAR REGRESSION

This study investigates the application of real-time Internet of Things (IoT) monitoring and predictive algorithms for optimizing liquid palm sugar production. By focusing on the prediction of Brix values, which indicate sugar concentration, the research aims to enhance process efficiency and product quality. Traditional manual methods of measuring Brix levels are often time-consuming and prone to inaccuracies. To address this, the study integrates IoT-based sensors that collect data on temperature, pressure, and weight during the evaporation process, using a linear regression model to predict Brix values in real time. Experimental results show that weight ratio-based predictions align well with manual refractometer readings, particularly in the early stages of production. However, deviations at higher Brix levels were noted, prompting the introduction of polynomial regression for improved accuracy. These findings suggest that IoT systems combined with predictive models offer a significant advancement in sugar production monitoring, reducing manual interventions and enhancing process control. The research contributes to the growing body of work on IoT applications in food production, particularly for liquid palm sugar processing, and provides a novel approach to addressing current challenges in Brix measurement.